Schizophrenia (SZ) is a severe and chronic psychiatric illness characterized by cognitive dysfunction. SZ has a higher prevalence in Veterans and is associated with increased disability and mortality in Veterans. Genome-wide association studies have identified variants of microglia-related genes as risk factors for SZ. How such genetic factors might be manifest as molecular alterations in microglia in SZ is not clear. Understanding microglial function in SZ is important because microglia are involved in the phagocytosis of dendritic spines on pyramidal neurons. Spines, which receive most of the excitatory input to pyramidal neurons, are critical mediators of the cognitive functions that are impaired in SZ. Spine density is lower principally in deep layer 3 of the prefrontal cortex (PFC) in SZ, and PFC layer 3 has been reported to subserve cognitive processes affected in the illness. Thus, we hypothesize that the dendritic spine deficit in deep layer 3 of the PFC in SZ is due to increased phagocytosis of spines by microglia. Testing our central hypothesis requires answering the following questions. First, do microglia in SZ show a profile of critical molecular features that enables increased spine phagocytosis (Aim 1)? We will quantify levels of both newly discovered and established transcripts that are selectively expressed by microglia and are involved in phagocytosis of dendritic spines (or the inhibition of phagocytosis) in laser microdissected deep layer 3 of PFC area 46 in SZ (n=32; 50% Veterans) and unaffected comparison subjects (n=32; 50% Veterans). We predict that SZ subjects have higher mRNA levels of microglia-specific markers that promote spine phagocytosis and lower mRNA levels of markers that inhibit phagocytosis. We will also use a novel quadruple-label RNAscope approach to quantify transcript levels in individual microglia to determine if all, or only some, microglia show molecular alterations in SZ. We predict that transcripts promoting spine phagocytosis are elevated in only a subset of microglia in SZ, which could account for findings that spine density is ~20% lower and most prominent in deep layer 3 of the PFC in SZ. Second, are spine deficits in deep layer 3 of the PFC in SZ associated with molecular and morphological features that indicate increased spine phagocytosis in nearby individual microglia (Aim 2)? Microglia and their associated processes have their own individual territorial organization, permitting the quantification of dendritic spines located exclusively within the territory of an individual microglia. Therefore, in the same subjects studied in Aim 1, we will use a quintuple-label confocal immunofluorescence microscopy approach to quantify the density of spines identified using two markers within 3D sampling regions constrained to the territorial domain of individual microglia. We will also quantify levels of CR3, which is involved in complement-mediated microglial phagocytosis of spines, and CD68, a phagocytic marker, and microglial process volume and complexity, which are associated with levels of spine phagocytosis, in individual microglia. We predict that spine density is inversely correlated with CR3 and CD68 levels and microglial process volume and complexity in nearby individual microglia in SZ subjects. Investigating the relationship between microglia-mediated spine phagocytosis and spine deficits in SZ will enable the pursuit of novel molecular targets for microglial inhibitors that may remediate dendritic spine deficits and repair synaptic connectivity, and possibly cognitive dysfunction, in Veterans with SZ.